Parser
Parser
Parser yra 2018 m. įkurta technologinių sprendimų ir konsultacijų įmonė, padedanti organizacijoms pritaikyti dirbtinį intelektą, duomenis ir programinės įrangos inžineriją sudėtingiems verslo poreikiams. Įmonės veikla apima individualių programinės įrangos sprendimų kūrimą, dirbtinio intelekto ir duomenų panaudojimą, produktų dizainą bei skaitmeninės transformacijos konsultacijas. Parser bendradarbiauja su technologijų lyderiais kurdama prie jų tikslų pritaikomus ir lengvai keičiamus sprendimus, derindama inžinerinę kompetenciją su praktinėmis inovacijomis ir ilgalaikėmis partnerystėmis su klientais.

Senior AI Engineer, Spain (Remote)

Build Python services, APIs, and production AI features for LLM-powered search evaluation and relevance tools. The role supports Parser’s multinational retail client.

Aprašymas

  • Develop AI tools that help Search teams diagnose issues, test changes, and measure relevance faster.
  • Create AI features that turn search behavior, diagnostics, and quality data into prioritized insights.
  • Build backend services and APIs to collect and combine human relevance judgments.
  • Convert the existing offline evaluation tool into on-demand environments developers can run in CI/CD.
  • Set up automated checks for AI feature accuracy and reliability, including LLM-based relevance judgments.
  • Partner with product, design, data science, and Search teams to bring prototypes into production.
  • Connect tools to existing client platforms and APIs.
  • Track quality, latency, and cost, and record key decisions and trade-offs.
  • Share progress early through documentation, demos, and incremental pull requests.
  • Deliver work in line with agreed architecture and engineering practices.
  • Clearly explain decisions, risks, and trade-offs.
  • Review AI-generated output quality with Search and data science colleagues.
  • Test assumptions against real data, then refine solutions based on evidence.
  • Move from exploration to delivery while keeping early solutions appropriately simple.
  • Build tools that users can trust, understand, and adopt.
  • Choose deterministic rules, traditional machine learning, or human review when they suit the task better than an LLM.

Reikalavimai

  • At least five years of professional experience in software engineering, data engineering, machine learning, or applied AI.
  • Practical experience building production LLM applications, including prompt design, structured outputs, and API integration.
  • Knowledge of embeddings, semantic and hybrid retrieval, and how they compare with lexical search.
  • Ability to assess models and approaches for quality, latency, security, and cost.
  • Experience creating offline evaluations and labelled-data benchmarks, including calibrating LLM judges against human labels.
  • Understanding of ranking measures such as nDCG, Precision at K, and MRR, as well as inter-rater agreement.
  • Awareness of hallucination, bias, drift, and privacy risks, with experience designing human review processes.
  • Strong SQL skills and experience working with large analytical datasets.
  • Familiarity with medallion architectures and connecting AI systems to data pipelines and APIs.
  • Strong Python skills, including experience developing maintainable backend services and APIs.
  • Sound testing practices for application, data, and AI components.
  • Experience with CI/CD, containers, and cloud infrastructure; Microsoft Azure is preferred.
  • Track record of delivering AI-enabled systems beyond standalone proofs of concept.
  • Ability to release iterative, user-focused solutions with appropriate reliability and safety measures.
  • Strong written and spoken communication, including explaining technical decisions and evaluation results to non-technical stakeholders.
  • Collaborative approach to delivering products across disciplines.
  • Preferred: Experience with retail or ecommerce search, including product taxonomy and long-tail queries.
  • Preferred: Practical experience with Elasticsearch or another production search platform, or with a vector database.
  • Preferred: Experience creating offline evaluation or A/B testing tools for search or recommendation.
  • Preferred: Experience developing human annotation or rating tools.
  • Preferred: Experience with event streaming such as Kafka, or with clickstream and web analytics data.

Privalumai

  • Highly competitive compensation package.
  • Medical insurance.
  • Flexible remote work.
  • Opportunity to join an organization experiencing triple-digit growth.
  • Opportunity to join a multicultural community of technology experts.

Susiję darbai

Gramian Consulting

Software Engineer for AI Code Licensing

Gramian Consulting

Submit an existing software repository for assessment and potential licensing. Compensation depends on source-code and Git history token counts and the agreed licensing terms.

Atidaryti
JUPUS

Senior Applied AI Engineer, RAG and Agentic Systems

JUPUS

Build production RAG pipelines and agentic workflows for JUPUS, a European legal technology company. Bring AI capabilities into legal case-file products used by lawyers.

Atidaryti
knowmad mood

Vyresnysis dirbtinio intelekto programinės įrangos architektas – nuotolinis darbas Ispanijoje

knowmad mood
1 001 – 5 000 Darbuotojai
KonsultavimasLogistikaSveikatos priežiūra

Vadovauk įmonių programinės įrangos architektūros ir generatyvinio dirbtinio intelekto iniciatyvoms bei konsultuok klientus. Prisidėk prie skaitmeninės transformacijos, taikomųjų programų modernizavimo ir dirbtinio intelekto mokymų įmonėje knowmad mood.

Atidaryti